jsmenu-codedev Mailing List for Morten's JavaScript Tree Menu (Page 2)
Status: Inactive
Brought to you by:
nettrom
You can subscribe to this list here.
2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(3) |
Dec
(5) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(10) |
Feb
(11) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
(1) |
Nov
|
Dec
|
From: Morten W. <war...@on...> - 2001-12-09 22:17:57
|
two new features added. also, I forgot to add the fact that MTMenuFrame had changed from a string to a pointer to the frame object in the new code. it's automagically set to point to the right frame in MTMGetFrames(). here's today's changelog: 2001-12-09: - added option MTMTrackedCookieName, used for storage of the tracked item in a cookie. MTMFetchCookie() is renamed to be plural, therefore it's now MTMFetchCookies(). the cookie is read on startup and if found MTMTrackedItem is set to the number in the cookie and the URL found is loaded into the found target frame. code run at startup is now quite patchy and should go through some redesign. the need for a URL resolver increases and a feature request has been opened. - added option MTMDefaultTarget (string) that lets you set a default target for new items. if it's not set the menu defaults back to the old usage of the empty string. - tested insertion of new mtmcode.js with new variables into old setup and noticed that it worked without errors. therefore it seems the addition of variables to mtmcode.js was successfull, making updates easier. Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |
From: Morten W. <war...@on...> - 2001-12-05 22:56:23
|
fixed a small bug, added some features that have been requested. started planning the "store tracked item in a cookie" feature, but didn't get around to implement it. logged entries: 2001-12-05: - fixed a bug in scrollbar colouring, one of the properties never got written to the document, and was therefore unsupported. - added configurable doctype for the menu document through MTMDOCTYPE - added configurable content-type for the menu document through MTMcontentType - added right-click prevention script through MTMrightClickMessage. currently unsupported in Mozilla/Netscape 6, one script I saw didn't work. will look for one that works and test it. - updated CREDITS to show credits for versions, as well as updated it with credits for some new features of 2.3.1 the changes are checked in to CVS, of course. Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |
From: Morten W. <war...@on...> - 2001-12-02 21:12:10
|
squashed two bugs, and did a slight usefulness update: 2001-12-02: - moved tracking code over to its own file, mtmtrack.js - changed MTMCheckURL() so it uses a for() loop and an array (excludeArray) thereby removing the ever-increasing if() statement. in the process I added a check for "file:" (so that protocol is also supported) and "/" so that URLs relative to the root directory are regarded as absolute. bugs #429698 and 423712 are therefore marked "closed - fixed". Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |
From: Morten W. <war...@on...> - 2001-11-25 20:20:20
|
as I just mentioned on the yahoogroups mailing list the menu has a back-button bug in IE6. from what I've been able to test IE6 doesn't support the 'replace' argument to document.open(), which the menu uses to make sure the rewriting of the menu frame doesn't add to the history list. the call to document.open() is done in line 205 in the CVS version: MTMDoc.open("text/html", "replace"); I'm not exactly sure how to go about fixing it, but I'm thinking along the lines of using the IE method insertAdjacentHTML(). the thing is, that when I first do things I prefer to do them well, so I am considering wanting to rewrite the frame usage so that the menu uses inserAdjacentHTML() to manipulate the document in IE6, and the W3C DOM methods in Netscape 6/Mozilla. I'm thinking along these lines: 1: have only one frame, it name is not important (but keeping it named 'code' is probably a good idea), containing code.html 2: in code.html load code that identifies the browser, just like it does now 3: if IE4+ or NN6, wait to the document is loaded, then use the appropriate dHTML methods to populate the document according to the user setup, showing the menu 4: if NN3, NN4 or Opera5/6, output a frameset code containing two frames, one for the code, one for the menu, with appropriate unique names. rename the parent frame to 'code' so the tracking code works, and make sure there's an object there pointing to MTMDisplayMenu() if this is done, then it should be possible to have the menu running without frames quite easily since the menu no longer needs the parent frameset to function (if we keep the HTML target-attribute usage). having an element in the document with an id of "mtmenu" that'll be populated by the code should be all. problem then is how to expand/collapse submenus, but that's a later issue. comments, anyone? Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |
From: Morten W. <war...@on...> - 2001-11-25 14:33:30
|
just a small update. after getting many requests about scroll bar colours I've added support for that. here's the changelog: 2001-11-25: - added scrollbar colour properties to code.html as well as support for this in mtmcode.js - changed the start-tag of the linked script element so it uses the IE-proprietary "defer" attribute. IE6 behaves unstable when linking a .js-file, and I believe telling it to not parse the linked script at runtime should help. More testing is needed. Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |
From: Morten W. <war...@on...> - 2001-11-23 23:47:02
|
it's high time this mailing list started getting some traffic (not to mention me re-reading "The Cathedral and the Bazaar" so I might start doing things right, but I digress...) anyway, started working on the code, here's what was done: 2001-11-23: - altered comment in mtmcode.js saying user-configurable options are in code.html - added section for default values of all user-configurable options - added variable MTMLinkedJSURL which can be used to link in a .js-file with own functions and variables to the menu document. tested in IE6, NN4.78 (Linux) and Mozilla 0.9.6 (Linux). needs more testing, and more documentation about method call quirks in browsers (onclick-events seem necessary at the moment) changes are checked into CVS. if anyone has any thoughts, ideas, comments or anything, feel free to start posting. I've got a bug list that's way too large, a bunch of ideas I never get 'round to implementing not to mention several user-submitted ideas & patches that should've made their way into the code many months ago... and if you want to tell me that I suck at project management that's ok too :) cheers, Morten! -- "...a liquid which was almost, but not quite, entirely unlike tea." My opinions are not necessarily those of my provider, not necessarily mine, and probably not necessary. |